#21  
Old 10-08-2015, 09:15 AM
SkyViking's Avatar
SkyViking (Rolf)
Registered User

SkyViking is offline
 
Join Date: Aug 2009
Location: Waitakere Ranges, New Zealand
Posts: 2,260
Quote:
Originally Posted by Paul Haese View Post
I have had the site tested with a DIMM a couple of times and seeing has been tested being as good as 0.7 arc seconds on a good night but more regularly at around 1.4 arc second. The drought years made the seeing very good.
Hi Paul,

I was thinking: How do you take advantage of 0.7" seeing when it appears at your site? Do you switch scope/camera systems?
I'm asking because for adequate sampling @ 0.7" FWHM you'd need an image scale of 0.25"/pixel (or even less).

I'm imaging at 0.86"/pixel (will be 0.75" now with Paracorr) which is reasonably sampled for my regular seeing conditions (2"-2.5"), but if it went down to 0.7" I'd be severely undersampling with my current setup and thus missing out on capturing the high resolution.

Quote:
Originally Posted by Paul Haese View Post
The FWHM I regularly get now is about 1.4-1.7" these days on both the wide and narrow field systems.
What's the image scale of your wide field system? Just thinking if it's wide field the image scale must be several arcseconds per pixel and thus how do you measure 1.4" FWHM in the wide field image?
Reply With Quote
  #22  
Old 10-08-2015, 03:47 PM
Slawomir's Avatar
Slawomir (Suavi)
Registered User

Slawomir is offline
 
Join Date: Sep 2014
Location: North Queensland
Posts: 3,240
Quote:
Originally Posted by codemonkey View Post
Definitely helps, thanks S! If it's not too much trouble (seriously, don't bother if it is), would you happen to know the median in each of those?
No problemo...

Median FWHM values in arcseconds for my 4" telescope:

Ha: 2.79 (24 subs collected over 3 different nights)
S2: 2.48 (also 24 subs over another 3 nights)
O3: 2.45 (16 subs, over 2 different nights)

8 nights in total so far...Yes, I know, it has been a slow process collecting data for this DSO...

Since eccentricity for all subs was in the range of 0.49-0.51 (in PixInsight eccentricity of zero is for perfectly round stars, and eccentricity as high as 0.42 is not noticable by most people) - there is some elongation to the stars, but consistent across all subs, indicating consistent tracking. I suspect tracking errors in RA axis to be the main cause.

So perhaps real seeing has been somewhat better than the above values suggest?
Reply With Quote
  #23  
Old 10-08-2015, 04:07 PM
codemonkey's Avatar
codemonkey (Lee)
Lee "Wormsy" Borsboom

codemonkey is offline
 
Join Date: Jul 2013
Location: Kilcoy, QLD
Posts: 2,058
Thanks S! :-)

Interesting results there; you're definitely doing better than I am in terms of median. I managed to get one broadband sub at ~2.4 last night which is maybe the best I've done to date.

Regarding eccentricity, I was under the impression that 0.5 meant round stars. Mine is usually 0.45-0.5 but I've not been paying much attention to it. Sounds like I need to look into that more.
Reply With Quote
  #24  
Old 10-08-2015, 04:37 PM
Slawomir's Avatar
Slawomir (Suavi)
Registered User

Slawomir is offline
 
Join Date: Sep 2014
Location: North Queensland
Posts: 3,240
Quote:
Originally Posted by codemonkey View Post
Thanks S! :-)

Interesting results there; you're definitely doing better than I am in terms of median. I managed to get one broadband sub at ~2.4 last night which is maybe the best I've done to date.

Regarding eccentricity, I was under the impression that 0.5 meant round stars. Mine is usually 0.45-0.5 but I've not been paying much attention to it. Sounds like I need to look into that more.
Hi Lee,

I think you might be getting slightly larger values because you use broadband filters, while my results are for 3nm filters? I think I will tweak my mount a bit in the coming weeks, as there is some backlash in both axes and see if that helps.

Yes, in PixInsight, eccentricity of zero indicates perfectly round stars, nonetheless, below 0.5 is still good
Reply With Quote
  #25  
Old 12-08-2015, 03:41 PM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Quote:
Originally Posted by SkyViking View Post
Hi Paul,

I was thinking: How do you take advantage of 0.7" seeing when it appears at your site? Do you switch scope/camera systems?
I'm asking because for adequate sampling @ 0.7" FWHM you'd need an image scale of 0.25"/pixel (or even less).

I'm imaging at 0.86"/pixel (will be 0.75" now with Paracorr) which is reasonably sampled for my regular seeing conditions (2"-2.5"), but if it went down to 0.7" I'd be severely undersampling with my current setup and thus missing out on capturing the high resolution.



What's the image scale of your wide field system? Just thinking if it's wide field the image scale must be several arcseconds per pixel and thus how do you measure 1.4" FWHM in the wide field image?
Hi Rolf.

I have two permanent systems in a separate observatories of their own to answer your first question.

You are right about not being able to take advantage of that sort of seeing. I don't want to go changing around cameras to suit the circumstances. It's just nice to have good seeing. It helps to make for sharp images, certainly sharper than some I see about. I could change the QSI over to the RC though and the STXL over to the FSQ. That would give me 0.46 per pixel on the RC. That would come closer than the current 0.76" per pixel.

Good point about the wide field system, I had not considered that the figures were smaller than the actual sampling. It being 2.01 seconds per pixel. I am just reading what I get in maxim, CCDstack and CCDinspector of FWHM values. Being that the maximum resolving power is 2 arc seconds per pixel that should feasibly mean I am sampling at the maximum resolution capable of the scope/camera combination and could not read any higher than that. Thoughts? I don't know why the programmes are showing these figures.
Reply With Quote
  #26  
Old 12-08-2015, 03:54 PM
Octane's Avatar
Octane (Humayun)
IIS Member #671

Octane is offline
 
Join Date: Dec 2005
Location: Canberra
Posts: 11,159
Can someone explain, in plain terms, what FWHM actually measures?

I believe I'm getting good figures (I'll post as soon as I get home from what the SubFrameSelector script reports for a bunch of stuff I've been getting lately), but, for a 3.5"/pixel configuration, what is a "good" value you'd expect to achieve?

This is an FSQ/STL combination.

Does MaxIm DL report it at capture time?

H
Reply With Quote
  #27  
Old 12-08-2015, 03:59 PM
codemonkey's Avatar
codemonkey (Lee)
Lee "Wormsy" Borsboom

codemonkey is offline
 
Join Date: Jul 2013
Location: Kilcoy, QLD
Posts: 2,058
Take the peak intensity of the star as recorded by your camera, then halve it. From there, inspect the star and find out how far away from the centroid that "half max" value is, and that'll give you your half width, half max.... to get the full width you need to look at the distance on both sides of the star. Hopefully that made some kind of sense? A picture would probably be easier...

Edit: I'm not sure what a "good" value would be with a 3.5"/px sampling rate. At 3.5"/px you're going to be very undersampled unless you have terrible, terrible seeing. Because of the impact undersampling has on stars (give them that square look with fast drop-off), I'm not sure that FWHM would be a good measure for you.

I have no idea about Maxim, I've never used it.
Reply With Quote
  #28  
Old 12-08-2015, 04:42 PM
Octane's Avatar
Octane (Humayun)
IIS Member #671

Octane is offline
 
Join Date: Dec 2005
Location: Canberra
Posts: 11,159
Cheers, Lee. I knew about the first bit. I'm just not sure what it objectively measures. That's what I'm having issues getting my head around.

I just ran a bunch of hydrogen alpha frames through SubFrameSelector:

1.339 - 2.39 with a median of 1.706.

I take it this isn't too bad?

H
Reply With Quote
  #29  
Old 12-08-2015, 06:00 PM
Slawomir's Avatar
Slawomir (Suavi)
Registered User

Slawomir is offline
 
Join Date: Sep 2014
Location: North Queensland
Posts: 3,240
Hi Humayun,

If I may intrude...

If the values you provided are in arcseconds then that would be VERY good.

But if these values are in pixels, then assuming 3.5"/pixel you are getting median FWHM of 6 arcseconds, Not as good as in the above scenario, but I suspect that when you are undersampling you will be naturally getting higher values of FWHM. Drizzle might help
Reply With Quote
  #30  
Old 12-08-2015, 06:01 PM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
Quote:
Originally Posted by Octane View Post
Cheers, Lee. I knew about the first bit. I'm just not sure what it objectively measures. That's what I'm having issues getting my head around.

I just ran a bunch of hydrogen alpha frames through SubFrameSelector:

1.339 - 2.39 with a median of 1.706.

I take it this isn't too bad?

H
if I may also intrude...

FWHM is a measure of star shape. It is a useful measure because all unsaturated stars (whatever brightness) in an unstretched image should have similar FWHM. I think that this characteristic is used by CCDinspector - if you have consistent FWHM, your field graph will be flat.

If you don't get consistent FWHM, you may be trying to measure on stretched data or your optics has problems. Also, some automated software will happily treat hot pixels as though they were stars and will also combine touching stars - the results are approximate at best. Better to use software that allows you to isolate one or more stars so that you can make sure that the test star(s) is not saturated and is not near a hot pixel or another star.

As Lee says, at 3.5 arcsec, you are heavily undersampled, so FWHM is probably not a very useful measure, because the star FWHMs will be similar to or smaller than the pixel size and what is reported will depend on where the stars fall on the pixel boundaries.

Last edited by Shiraz; 13-08-2015 at 07:36 AM.
Reply With Quote
  #31  
Old 12-08-2015, 06:12 PM
Octane's Avatar
Octane (Humayun)
IIS Member #671

Octane is offline
 
Join Date: Dec 2005
Location: Canberra
Posts: 11,159
Hi Slawomir,

Those values are indeed in arcseconds, as taken directly from PixInsight's SubFrameSelector script's output. I make use of Drizzle on my binned RGB data. It has worked incredibly well for me!

Hi Ray,

Thanks for the explanation! I understand that I'm heavily undersampled at 3.5"/pixel. Even if it's not a truly useful measure in my case, do the values I've quoted (which are actually quite typical, across my data) mean that I have reasonable seeing in my location? Mind you, I only image between 45 degrees either side of zenith. This isn't even in my control, as I've specifically told my automation software to only image between those altitudes.

H
Reply With Quote
  #32  
Old 12-08-2015, 06:27 PM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
Quote:
Originally Posted by Octane View Post
Hi Slawomir,

Those values are indeed in arcseconds, as taken directly from PixInsight's SubFrameSelector script's output. I make use of Drizzle on my binned RGB data. It has worked incredibly well for me!

Hi Ray,

Thanks for the explanation! I understand that I'm heavily undersampled at 3.5"/pixel. Even if it's not a truly useful measure in my case, do the values I've quoted (which are actually quite typical, across my data) mean that I have reasonable seeing in my location? Mind you, I only image between 45 degrees either side of zenith. This isn't even in my control, as I've specifically told my automation software to only image between those altitudes.

H
As Slawomir points out, about the best you can say is that the seeing is probably better than 6 arcsec, which is not saying much. It could of course be much better than that, but by undersampling, you cannot tell.

This is one major advantage of undersampled imaging - you can get the same results whatever the seeing is like.
Reply With Quote
  #33  
Old 12-08-2015, 06:31 PM
Octane's Avatar
Octane (Humayun)
IIS Member #671

Octane is offline
 
Join Date: Dec 2005
Location: Canberra
Posts: 11,159
Thanks, that makes sense.

Now, to find a 1"/pixel setup to gauge my seeing!

H
Reply With Quote
  #34  
Old 13-08-2015, 06:13 AM
Atmos's Avatar
Atmos (Colin)
Ultimate Noob

Atmos is offline
 
Join Date: Aug 2011
Location: Melbourne
Posts: 6,980
Quote:
Originally Posted by Octane View Post
Thanks, that makes sense.

Now, to find a 1"/pixel setup to gauge my seeing!

H
TeleVue 4x Powermate
Reply With Quote
  #35  
Old 13-08-2015, 06:30 AM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
One thing you could try Paul is more aggressive autoguiding on nights of better seeing. Aggressiveness 10 and min move even lower. That was advice from Roland Christen.

Greg.


Quote:
Originally Posted by Paul Haese View Post
Hi Rolf.

I have two permanent systems in a separate observatories of their own to answer your first question.

You are right about not being able to take advantage of that sort of seeing. I don't want to go changing around cameras to suit the circumstances. It's just nice to have good seeing. It helps to make for sharp images, certainly sharper than some I see about. I could change the QSI over to the RC though and the STXL over to the FSQ. That would give me 0.46 per pixel on the RC. That would come closer than the current 0.76" per pixel.

Good point about the wide field system, I had not considered that the figures were smaller than the actual sampling. It being 2.01 seconds per pixel. I am just reading what I get in maxim, CCDstack and CCDinspector of FWHM values. Being that the maximum resolving power is 2 arc seconds per pixel that should feasibly mean I am sampling at the maximum resolution capable of the scope/camera combination and could not read any higher than that. Thoughts? I don't know why the programmes are showing these figures.
Reply With Quote
  #36  
Old 13-08-2015, 08:22 AM
rustigsmed's Avatar
rustigsmed (Russell)
Registered User

rustigsmed is offline
 
Join Date: Mar 2012
Location: Mornington Peninsula, Australia
Posts: 3,950
I threw sharper subs that didn't have my tilt issues into DSS to see what the FWHM rating was from a variety of times throughout the past 6 months most of them were between 5-6.5. the best batch I saw was in the mid to high 3s but that was with an Oiii filter. all with a dslr and yes good old Melbourne
Reply With Quote
  #37  
Old 13-08-2015, 09:04 AM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,937
Quote:
Originally Posted by Shiraz View Post

As Lee says, at 3.5 arcsec, you are heavily undersampled, so FWHM is probably not a very useful measure, because the star FWHMs will be similar to or smaller than the pixel size and what is reported will depend on where the stars fall on the pixel boundaries.
In that case, am I to assume that reported values for my wide field system are inaccurate too? It might be prudent to obtain one of these. Just don't fall off the chair when you see the price.
Reply With Quote
  #38  
Old 13-08-2015, 10:24 AM
SkyViking's Avatar
SkyViking (Rolf)
Registered User

SkyViking is offline
 
Join Date: Aug 2009
Location: Waitakere Ranges, New Zealand
Posts: 2,260
Quote:
Originally Posted by Paul Haese View Post
In that case, am I to assume that reported values for my wide field system are inaccurate too? It might be prudent to obtain one of these. Just don't fall off the chair when you see the price.
Hi Paul, I'd say yes FWHM as a measurement of seeing only makes sense when images are properly oversampled. If the stars are less than a couple of pixels wide then measurement is impossible.
To give an example, when I look at my oversampled raw frames I can see stars typically covering 5-6 pixels across, with 2-3 significantly brighter pixels at the very centre. By definition these 2-3 pixels make up the FWHM. Hence Maxim's reported FWHM of ~2 arcseconds makes perfect sense given this equals 2-3 pixels at my image scale of 0.86"/px.
However, if I inspect a 2x2 binned raw frame taken at the same time in the same seeing conditions I get nonsensical FWHM values from Maxim because the stars are now undersampled and so the tool cannot measure properly.

You mentioned using a DIMM to measure seeing, but as I understand it this is done at high frequency and therefore not applicable to deep sky imaging? It seems more akin to planetary imaging where the details that can be teased out with lucky imaging are of course much finer.

For deep sky purposes I think the only reliable measurement is actual FWHM of stars in subframes with integration times of tens of seconds or more.
Reply With Quote
  #39  
Old 13-08-2015, 12:31 PM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
Quote:
Originally Posted by SkyViking View Post

You mentioned using a DIMM to measure seeing, but as I understand it this is done at high frequency and therefore not applicable to deep sky imaging? It seems more akin to planetary imaging where the details that can be teased out with lucky imaging are of course much finer.

For deep sky purposes I think the only reliable measurement is actual FWHM of stars in subframes with integration times of tens of seconds or more.
Hi Rolf. The DIMM certainly does measure image motion at high speed, but only to get a full bandwidth measure of the effect of the turbulence. The long term seeing FWHM applicable to DSO imaging is then obtained by plugging the DIMM output into the standard seeing model - nicely summarised in the reference.

Measured image FWHM is a fair estimate of seeing, but it also includes mount tracking and guide wobble effects that are automatically removed by a DIMM - these effects are convolved with the atmospheric seeing when imaging, so image FWHM is worse than that from seeing alone. DIMM seems to be the gold standard at present.

ref: http://www.ctio.noao.edu/~atokovin/papers/pasp2002.pdf

Last edited by Shiraz; 13-08-2015 at 03:07 PM.
Reply With Quote
  #40  
Old 13-08-2015, 05:12 PM
SkyViking's Avatar
SkyViking (Rolf)
Registered User

SkyViking is offline
 
Join Date: Aug 2009
Location: Waitakere Ranges, New Zealand
Posts: 2,260
Quote:
Originally Posted by Shiraz View Post
Hi Rolf. The DIMM certainly does measure image motion at high speed, but only to get a full bandwidth measure of the effect of the turbulence. The long term seeing FWHM applicable to DSO imaging is then obtained by plugging the DIMM output into the standard seeing model - nicely summarised in the reference.

Measured image FWHM is a fair estimate of seeing, but it also includes mount tracking and guide wobble effects that are automatically removed by a DIMM - these effects are convolved with the atmospheric seeing when imaging, so image FWHM is worse than that from seeing alone. DIMM seems to be the gold standard at present.

ref: http://www.ctio.noao.edu/~atokovin/papers/pasp2002.pdf
Hi Ray, thanks for the info re DIMM, I was not aware of the details.

Yes you are absolutely right that measured FWHM includes contributions from other sources. I was just thinking that in order to compare results we must measure the same thing, and nearly all replies on this thread has referenced FWHM as measured directly on images.
The actual seeing is certainly better than what we can record on long exposures, but not many have DIMM or other such equipment to measure that.
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 02:01 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Testar
Advertisement